Methane as an Arctic

Arctic Monitoring and Assessment Programme (AMAP)

AMAP Assessment 2015:

Methane as an Arctic

climate forcer

freely as teaching materials and for other educational purposes.

The only condition of such use is acknowledgement of AMAP as the source of the material according to the recommended citation.

In case of questions regarding educational use, please contact the AMAP Secretariat (amap@amap.no).

Note: This report may contain material (e.g. photographs) for which permission for use will need to be obtained from original copyright holders.

Disclaimer: The views expressed in this peer-reviewed report are the responsibility of the authors of the report and do not necessarily reflect the views of the Arctic Council, its members or its observers.

Arctic Monitoring and Assessment Programme (AMAP) Oslo, 2015

AMAP Assessment 2015:

Methane as an Arctic

climate forcer

AMAP Assessment 2015: Methane as an Arctic climate forcer

Citation

AMAP Assessment 2015: Methane as an Arctic climate forcer. Arctic Monitoring and Assessment Programme (AMAP), Oslo, Norway. vii + 139 pp.

ISBN – 978-82-7971-091-2

© Arctic Monitoring and Assessment Programme, 2015 Published by

Arctic Monitoring and Assessment Programme (AMAP), Oslo, Norway (www.amap.no) Ordering

This report can be ordered from the AMAP Secretariat, Gaustadalléen 21, N-0349 Oslo, Norway This report is also published as electronic documents, available from the AMAP website at www.amap.no

Production

Production management

Simon Wilson (AMAP Secretariat) Scientific, technical and linguistic editing

Carolyn Symon (carolyn.symon@btinternet.com) Lay-out and technical production

Burnthebook, United Kingdom (www.burnthebook.co.uk) Design and production of computer graphics

Simon Duckworth (simon@burnthebook.co.uk) Cover photograph

An ecologist deploys a gas trap to collect methane bubbling up from the bottom of a frozen lake, Alaska.

Photo: © Mark Thiessen/National Geographic Creative/Corbis Printing

Narayana Press, Gylling, DK-8300 Odder, Denmark (www.narayanapress.dk)

AMAP Working Group (during the period of preparation of this assessment)

Morten Olsen (Chair, Denmark), Russel Shearer (Vice-Chair, Canada), Fred Wrona (Canada), Mikala Klint (Denmark), Outi Mähönen (Vice-chair, Finland), Helgi Jensson (Iceland), Per Døvle (Norway), Tove Lundberg (Sweden), Yuri Tsaturov (Vice-chair, Russia), Tom Armstrong (USA)

AMAP Secretariat

Lars-Otto Reiersen, Simon Wilson, Jon Fuglestad, Jan-Rene Larsen, Janet Pawlak, Inger Utne Arctic Council Member States and Permanent Participants of the Council:

Canada, Denmark/Greenland/Faroe Islands, Finland, Iceland, Norway, Russia, Sweden, United States, Aleut International Association (AIA), Arctic Athabaskan Council (AAC), Gwitch’in Council International (GCI), Inuit Circumpolar Council (ICC), Russian Association of Indigenous Peoples of the North (RAIPON), Saami Council

Acknowledgments

Authors

Vivek K. Arora (Canada), Terje Berntsen (Norway), Arne Biastoch (Germany), Philippe Bousquet (France), Lori Bruhwiler (USA), Elizabeth Bush (Canada), Elton Chan (Canada), Torben R. Christensen (Sweden and Denmark), Ed Dlugokencky (USA), Rebecca E. Fisher (UK ), James France (UK), Michael Gauss (Norway), Lena Höglund-Isaksson (Sweden and Austria), Sander Houweling (The Netherlands), Kovan Huissteden (The Netherlands), Greet Janssens-Maenhout (Italy), Anna Karion (USA), Andrei Kiselev (Russia), Charles D. Koven (USA), Kerstin Kretschmer (Germany), Kaarle Kupiainen (Finland), J. Michael Kuperberg (USA), Tuomas Laurila (Finland), Guilong Li (Canada), David Lowry (UK), Joe Melton (Canada), John Miller (USA), Euan G.

Nisbet (UK), Dirk J.L. Olivié (Norway), Giuliana Panieri (Norway ), Frans-Jan W. Parmentier (Sweden), David A. Plummer (Canada), Shilpa Rao (Norway and Austria), Torsten Sachs (Germany), Marjorie Shepherd (Canada), Anna Silyakova (Norway), O. Amunde Søvde (Norway ), Colm Sweeney (USA), Carrie Taylor (Canada), Allison Thomson (USA), James W.C. White (USA), Douglas Worthy (Canada), Wenxin Zhang (Sweden)

Bold: Coordinating Chapter Authors

Reviewers

Reino Abrahamsson (Sweden), P. Bergamaschi (Italy), Terje Bernsten (Norway), Dominique Blain (Canada), J. Butler (USA), Patrick Crill (Sweden), Jordan Guthrie (Canada), Paal Kolka Jonsson (Iceland), Jennifer Kerr (Canada), Dave McGuire (USA), Helge Niemann (Germany), Michael Prather (USA), Pieter Tans (USA), Apostolos Voulgarakis (UK), John Walsh (USA), Oliver Wild (UK), Christoph Wöll (Iceland)

Contents

Acknowledgments

Preface

1.

Introduction

1.1 Background . . . . 1

1.2 The Arctic climate context: Past and future warming . . . . 3

1.3 Report structure . . . . 3

2.

The global methane budget and the role of methane in climate forcing

2.1 Background . . . . 7

2.2 Overview of natural and anthropogenic methane sources . . . . 7

2.3 Overview of methane sinks . . . . 9

2.4 Methane and the hydroxyl radical . . . . 10

2.4.1 Observation-based estimates of hydroxyl . . . . 12

2.4.2 Photochemical modelling estimates of hydroxyl . . . . 13

2.4.3 Long-term changes in hydroxyl . . . . 13

2.5 Methane radiative forcing . . . . 14

3.

Natural terrestrial methane sources in the Arctic

3.1 Introduction . . . . 15

3.2 Description of natural terrestrial methane sources . . . . 15

3.2.1 Processes . . . . 15

3.3 Methods for measuring methane fluxes . . . . 18

3.4 Quantification of methane emissions from Arctic terrestrial sources . . . . 20

3.5 Quantification of carbon stocks and spatial extent . . . . 21

3.5.1 Size and characteristics of the Arctic soil carbon reservoir . . . . 21

3.5.2 Vulnerability of the Arctic soil carbon reservoir . . . . 23

3.6 Estimates of future methane emissions from natural terrestrial sources . . . . 24

3.7 Natural terrestrial emissions for AMAP methane climate modeling . . . . 24

3.8 Conclusions . . . . 25

3.8.1 Key findings . . . . 25

3.8.2 Recommendations . . . . 25

4.

Natural marine methane sources in the Arctic

4.1 Introduction . . . . 27

4.2 Methane sources and reservoirs in the Arctic Ocean . . . . 28

4.2.1 Subsurface methane production . . . . 28

4.2.2 Gas hydrate formation and occurrence . . . . 29

4.2.3 Surface water sources of methane . . . . 30

4.3 Controls on methane sources . . . . 30

4.3.1 Anaerobic and aerobic consumption of methane . . . . 30

4.3.2 Fate of rising methane bubbles . . . . 31

4.4 Emission to the atmosphere . . . . 31

4.4.1 Measurement techniques . . . . 31

4.4.2 Arctic Ocean emission estimates . . . . 32

4.5 Evidence of methane release in the geologic past . . . . 32

4.6 Hydrate modeling . . . . 33

4.6.1 Modeling rationale . . . . 33

4.6.2 Model setup . . . . 34

4.6.3 Hydrate abundance and vulnerability to warming . . . . 34

4.7 Estimates of future Arctic Ocean emissions . . . . 36

4.7.1 Deep water gas hydrate deposits . . . . 36

4.7.2 Subsea permafrost . . . . 36

4.7.3 Ocean surface . . . . 37

4.8 Conclusions . . . . 37

4.8.1 Key findings. . . . 37

4.8.2 Recommendations. . . . 38

5.

Anthropogenic methane sources, emissions and future projections

5.1 Introduction . . . . 39

5.2 Global anthropogenic methane emissions in past years . . . . 39

5.2.1 Emission inventory approach . . . . 39

5.2.2 Sources of uncertainty in methane emissions estimates . . . . 39

5.2.3 Recent inventories of global anthropogenic methane emissions . . . . 40

5.2.4 Global methane emissions from oil and natural gas systems . . . . 43

5.3 Global projections of future anthropogenic methane emissions . . . . 46

5.3.1 Use of integrated assessment models in climate policy . . . . 46

5.3.2 Global baseline and mitigation scenarios for anthropogenic methane emissions . . . . 47

5.3.3 Global technical abatement potential for methane by technology . . . . 49

5.3.4 Future emissions and technical reduction potentials by world region . . . . 50

5.3.5 Cost of future reductions in global anthropogenic methane emissions . . . . 52

5.4 Anthropogenic methane emissions in Arctic nations . . . . 53

5.4.1 Contribution of Arctic nations to current and future anthropogenic methane emissions . . . . 53

5.4.2 Sources and abatement potentials for anthropogenic methane emissions in Arctic nations . . . . 53

5.4.3 Sources and abatement potentials for anthropogenic methane emissions by country . . . . 54

5.4.4 Uncertainty in oil and gas systems emissions in Arctic nations . . . . 57

5.5 Use of anthropogenic methane emission scenarios in climate models . . . . 58

5.6 Conclusions . . . . 58

5.6.1 Key findings . . . . 58

5.6.2 Recommendations . . . . 59

6.

Long-term monitoring of atmospheric methane

6.1 Introduction . . . . 61

6.2 Surface observations of atmospheric methane . . . . 61

6.3 Large-scale trends in Arctic atmospheric methane . . . . 63

6.4 Continuous methane measurements at Arctic locations . . . . 65

6.4.1 Diurnal and day-to-day variability . . . . 65

6.4.2 Seasonal and interannual variability . . . . 67

6.4.3 Trajectory cluster analysis . . . . 67

6.5 Methane measurements at Tiksi on the coast of the Laptev Sea . . . . 69

6.6 Isotopic measurements . . . . 72

6.6.1 Available data . . . . 72

6.6.2 Annual cycle . . . . 73

6.6.3 Identification of Arctic methane sources . . . . 74

6.7 Conclusions . . . . 74

6.7.1 Key findings . . . . 74

6.7.2 Recommendations . . . . 75

7.

Modeling of atmospheric methane using inverse (and forward) approaches

7.1 Introduction . . . . 77

7.2 Inverse modeling approaches for understanding Arctic methane emissions. . . . 77

7.2.1 Introduction to atmospheric inverse modeling . . . . 77

7.2.2 Role of uncertainty in inverse modeling . . . . 79

7.2.3 Importance of adequate observational coverage. . . . 79

7.2.4 Results from inverse model studies . . . . 80

7.3 Evaluating global wetland models using forward modeling and atmospheric observations . . . . 84

7.3.1 Evaluation of wetland models – methods . . . . 85

7.3.2 Evaluation of wetland models – results . . . . 86

7.4 Conclusions . . . . 87

7.4.1 Key findings . . . . 87

7.4.2 Recommendations . . . . 88

Appendix: Global atmospheric inverse model studies reviewed for the comparison discussed in Section 7.2.4. . . . . 89

8.

Modeling the climate response to methane

8.1 Introduction . . . . 91

8.2 Climate effects of historical changes in methane concentration . . . . 92

8.3 Effects of changing anthropogenic and natural methane emissions . . . . 93

8.3.1 Box model – Earth System Model calculations . . . . 93

8.3.2 Chemical Transport Model – Earth System Model calculations . . . . 99

8.4 Conclusions . . . . 103

8.4.1 Key findings . . . .103

8.4.2 Recommendations . . . .105

Appendix: One-box model of atmospheric methane . . . .106

9.

Conclusions and Recommendations

9.1 Context . . . . 107

9.2 Recommendations for research and monitoring . . . . 110

9.2.1 Natural emissions . . . .110

9.2.2 Anthropogenic emissions . . . .111

9.2.3 Observations and inverse modeling . . . .111

9.2.4 Earth system modeling . . . .111

9.3 Final comments . . . . 111

Annex: Modeling the climate response – A summary

A1 Introduction . . . . 113

A2 Modeling approach . . . . 113

A2.1 ^VSLCFs . . . .113

A2.2 ^SLCFs . . . .113

A3 Summary of main results . . . . 113

A4 Results from the Expert Group on Black Carbon and Ozone . . . . 114

A4.1 ^Ozone . . . .114

A4.2 Results from the ECLIPSE transient simulations . . . .115

A5 Results from the Expert Group on Methane . . . . 116

References

Acronyms and abbreviations

Preface

This assessment report presents the results of the 2015 AMAP Assessment of Methane as an Arctic climate forcer. This is the first AMAP assessment dealing with this issue and complements a second assessment of black carbon and tropospheric ozone as Arctic climate forcers.

The Arctic Monitoring and Assessment Programme (AMAP) is a group working under the Arctic Council. The Arctic Council Ministers have requested AMAP to:

• produce integrated assessment reports on the status and trends of the conditions of the Arctic ecosystems;

• identify possible causes for the changing conditions;

• detect emerging problems, their possible causes, and the potential risk to Arctic ecosystems including indigenous peoples and other Arctic residents; and to

• recommend actions required to reduce risks to Arctic ecosystems.

This report provides the accessible scientific basis and validation for the statements and recommendations made in the Summary for Policy-makers: Arctic Climate Issues  reportⁱ that was delivered to Arctic Council Ministers at their meeting in Iqaluit, Canada in April . It is also the basis for a related AMAP State of the Arctic Environment report Arctic Climate Issues

: Overviewⁱⁱ . It includes extensive background data and references to the scientific literature, and details the sources for figures reproduced in the overview report. Whereas the Summary for Policy-makers report contains recommendations that focus mainly on policy-relevant actions concerned with addressing short-lived climate forcers, the conclusions and recommendations presented in this report also cover issues of a more scientific nature, such as proposals for filling gaps in knowledge, and recommendations relevant to future monitoring and research work.

This assessment of methane as an Arctic climate forcer was conducted between 2012 and 2014 by an international group of over 40 experts. Lead authors were selected based on an open nomination process coordinated by AMAP. A similar process was used to select international experts who independently reviewed this report.

Information contained in this report is fully referenced and based first and foremost on peer-reviewed and published results of research and monitoring undertaken since 2010.

It also incorporates some new (unpublished) information from monitoring and research conducted according to well- established and documented national and international standards and quality assurance/quality control protocols. Care has been taken to ensure that no critical probability statements are based on non-peer-reviewed materials.

Access to reliable and up-to-date information is essential for the development of science-based decision-making regarding ongoing changes in the Arctic and their global implications.

The methane assessment summary reports^{i, ii} have therefore been developed specifically for policy-makers, summarizing the main findings of the assessment. The methane assessment lead authors have confirmed that both this report and its derivative products accurately and fully reflect their scientific assessment.

The methane assessment reports are freely available from the AMAP Secretariat and on the AMAP website: www.amap.no, and their use for educational purposes is encouraged.

AMAP would like to express its appreciation to all experts who have contributed their time, efforts and data, in particular the lead authors who coordinated the production of this report.

Thanks are also due to the reviewers who contributed to the methane assessment peer-review process and provided valuable comments that helped to ensure the quality of the report. A list of contributors is included in the acknowledgements at the start of this report and lead authors are identified at the start of each chapter. The acknowledgements list is not comprehensive.

Specifically, it does not include the many national institutes, laboratories and organizations, and their staff, which have been involved in various countries in methane-related monitoring and research. Apologies, and no lesser thanks are given to any individuals unintentionally omitted from the list.

The support from the Arctic countries and non-Arctic countries implementing research and monitoring in the Arctic is vital to the success of AMAP. The AMAP work is essentially based on ongoing activities within these countries, and the countries that provide the necessary support for most of the experts involved in the preparation of the AMAP assessments. In particular, AMAP would like to acknowledge Canada and the United States for taking the lead country role in this assessment and thank Canada, Norway and the Nordic Council of Ministers for their financial support to the methane assessment work.

The AMAP Working Group is pleased to present its assessment to the Arctic Council and the international science community.

Marjorie Shepherd (Methane Assessment Co-lead, Canada) J. Michael Kuperberg (Methane Assessment Co-lead, USA) Morten Olsen (AMAP Chair, April 2015)

Lars-Otto Reiersen (AMAP Executive Secretary) Oslo, September 2015

i. AMAP, 2015. Summary for Policy-makers: Arctic Climate Issues 2015. Arctic Monitoring and Assessment Programme (AMAP), Oslo, Norway. 16 pp.

ii. AMAP, 2015. Climate Issues 2015: Overview report. Arctic Monitoring and Assessment Programme (AMAP), Oslo, Norway. 16 pp.

1. Introduction

A: E B, M K, M S

1.1

Background

This chapter sets out the context and motivation for undertaking this assessment of methane and the Arctic climate, and provides a guide for readers to the chapters that follow. The overarching context for this assessment is the concern of Arctic nations for the consequences of the large and rapid changes in regional climate that are already underway – evident in observational records and projected to continue. This concern has resulted in two major assessments of Arctic climate change under the auspices of Arctic Council: The Arctic Climate Impact Assessment (ACIA 2005) and Snow, Water, Ice and Permafrost in the Arctic (SWIPA): Climate Change and the Cryosphere (AMAP 2011a).

These assessments documented the widespread changes already occurring across the physical landscapes and ecosystems of the Arctic, and highlighted risks associated with the projected continuation and potential acceleration of observed changes if anthropogenic drivers of Arctic warming continue. An anthropogenic contribution to Arctic warming over the last 50 years has been established (Bindoff et al. 2013) and future scenarios of Arctic and global climate change generally assume additional emissions of anthropogenic greenhouse gases, although these emissions vary in timing and magnitude across scenarios. The challenges that future climate-related risks present to people living in the Arctic were also comprehensively described in the aforementioned Arctic assessment reports.

Reducing the rate and magnitude of Arctic warming during this century will require global comprehensive strategies to address the suite of greenhouse gases and other substances driving anthropogenic climate change. As is clear from Fig. 1.1 (Collins et al. 2013), carbon dioxide will dominate radiative forcing under a range of future scenarios, as it does currently, accounting for about 80–90% of total anthropogenic forcing in the year 2100. Carbon dioxide is the main persistent (long- lived¹) greenhouse gas contributing to anthropogenic climate change and it is now well established that total cumulative emissions of carbon dioxide are the main determinant of long- term global warming (Collins et al. 2013). Therefore, reducing emissions of carbon dioxide is the backbone of any meaningful effort to limit global and Arctic warming. However, emissions of other substances also contribute substantially to present-day radiative forcing and as long as such emissions continue, these substances will continue to contribute to total radiative forcing and thus to global warming and associated climate changes (Fig. 1.1). Comprehensive climate-change mitigation would encompass strategies to reduce emissions of all climate forcing agents. Any ongoing anthropogenic methane emissions, for example, would elevate climate warming above that induced

by carbon dioxide alone. Reducing emissions of methane (with an atmospheric lifetime of about a decade; see Ch. 2) and other substances with shorter atmospheric lifetimes than carbon dioxide (e.g. from days to decades) provides an opportunity to reduce radiative forcing in the near term (i.e. in the years immediately following the reduction in emissions) since atmospheric concentrations of short-lived substances can be lowered more quickly through emission reductions. The timescale of the response, in terms of lowering atmospheric concentrations, is dependent on the atmospheric lifetime of the particular substance (see Ch. 2). Indeed, it has already been shown that reducing emissions of methane and black carbon can help reduce projected global and Arctic warming in the near term (UNEP and WMO 2011; Shindell et al. 2012) and this work provided a foundation for the Arctic Council to initiate some targeted work directed towards understanding the role of short-lived climate forcers in Arctic climate and the potential benefits of mitigating such substances.

The Arctic Council, through its Arctic Monitoring and Assessment Programme (AMAP) working group, has undertaken scientific work directed at understanding the role of short-lived climate forcers in Arctic climate change, targeting black carbon, ozone and methane for focused study². It is in this context that the Short-Lived Climate Forcers Expert Group on Methane (henceforth referred to as the Methane Expert Group) was established and tasked by AMAP to provide scientific information to inform methane mitigation planning by Arctic nations. The intent of this work is to better understand the contribution methane mitigation can make to reducing the rate of Arctic warming in the near term. Methane is a globally well-mixed greenhouse gas meaning levels of methane in the atmosphere are similar around the globe. In turn, this means that reductions in emissions anywhere contribute to reducing atmospheric methane levels. Assessing methane mitigation benefits will provide information about the ability of Arctic nations to influence methane levels, putting potential mitigation by Arctic nations in context with global methane mitigation potential.

Atmospheric methane levels are, however, influenced by natural as well as anthropogenic emission sources. Therefore, whether or not Arctic (and global) methane concentrations can be lowered depends not only on mitigation of anthropogenic methane sources but also on future changes in natural methane sources. In the Arctic, rising temperatures have the potential to enhance the release of methane from natural sources.

There are known to be very large reservoirs of methane and organic carbon in the Arctic Ocean seabed, and on land in the

1 Carbon dioxide is removed from the atmosphere through a variety of biogeochemical processes operating on different timescales. Some portion of emitted carbon dioxide stays in the atmosphere for millennia (Ciais et al. 2013).

2 AMAP established a Short-Lived Climate Forcer Expert Group in 2009. Initially, this group focused on understanding the role of black carbon in Arctic climate and produced an assessment report in 2011 (AMAP 2011b). The first expert group is now focusing on both black carbon and ozonewhile a second expert group, focusing on methane, has been established.

soils and lake sediments of the Arctic. Decomposition of the carbon in these reservoirs can lead to the emission of either carbon dioxide or methane, depending on the conditions under which decomposition occurs, with wet, anaerobic conditions favoring methane production (see Ch. 3). These reservoirs are contained (completely or partially) by ice, frozen soil and frozen sediment, and the surface–air exchange is partially mediated by ice cover.

The Arctic Council led SWIPA report (AMAP 2011a; Callaghan et al. 2011) and other recent major scientific assessments (Ciais et al. 2013; NRC 2013) drew attention to the risks associated with warming, thawing and destabilizing of land and subsea permafrost in the Arctic, particularly with regard to the potential for enhanced fluxes of methane to the atmosphere. As methane is a powerful greenhouse gas, and given the sheer size of Arctic carbon reservoirs, there is a recognized potential for climatically significant methane emissions from the Arctic, which would represent a positive, amplifying feedback on the global climate system (i.e. warming increases methane emissions which in turn drive further warming…, and so on). While recent comprehensive reviews of the published literature have concluded that gradual rather than abrupt increases in Arctic methane emissions over the 21st century are more likely, with a moderate positive feedback on climate (Ciais et al. 2013; NRC 2013), these same studies also emphasized that scientific understanding of the topic is immature with many uncertainties about the controlling processes and timescales for enhanced methane releases from the ocean and land.

In recognition of the status of this area of science as an emerging issue, the Methane Expert Group was also tasked to provide an in-depth report on potential future increases in methane from natural sources in the Arctic in response to projected regional warming. While such emissions are not under the direct control of Arctic nations, understanding the contributions of these sources to changing atmospheric methane levels will influence Arctic nations’ ability to attribute reductions in atmospheric methane levels to any anthropogenic methane mitigation they, or other countries, may undertake.

This document is a report on the work completed by the Methane Expert Group in the two and a half years since its inception. The report is developed from a review of relevant literature as well as targeted scientific analyses designed to help fulfill the mandate of providing policy-relevant science advice to the Arctic Council, through AMAP, to inform discussions of actions on short-lived climate forcers. This work was directed at answering two major questions:

What is the potential benefit, in terms of reduced Arctic warming, of methane emissions mitigation by Arctic nations?

How does the magnitude of potential emission reductions from anthropogenic sources compare to potential changes in methane emissions from natural sources in the Arctic?

Clearly, future Arctic warming will be influenced not only by actions taken by Arctic nations to reduce anthropogenic emissions of methane and other climate forcers, but also by actions taken by the rest of the world. Therefore, it is important to understand the potential benefit of methane mitigation by Arctic nations in this larger context. While the methane mitigation potential of Arctic nations is put in context with global methane mitigation potential in this report, an integrated consideration of the benefits of mitigation measures in a multi- pollutant framework was outside the mandate of the Methane Expert Group. Similarly, it is recognized that there is also a larger context or backdrop to the question about natural methane emissions addressed by the group as major natural sources of methane also exist outside the Arctic, most notably in the tropics. Changes in natural methane emissions both within and outside the Arctic will respond to climate changes that ensue in response to total radiative forcing changes, driven primarily by carbon dioxide (see earlier). The focus on methane mitigation and on Arctic methane sources in this report reflects the task given to the Methane Expert Group and the expertise among its members.

3 The four RCPs are described in Ch. 5. In brief, these four scenarios were the basis for future climate change projections under the Coupled Model Intercomparison Project Phase 5 (CMIP5), with these projections featured in the Working Group I contribution to the IPCC Fifth Assessment (Collins et al. 2013c). The four RCPs span a range of potential future radiative forcing, from a high emission scenario (RCP8.5) to one that aims to limit global warming to about 2°C (RCP2.6).

Fig. 1.1 The left-hand panel shows the contribution of individual anthropogenic forcings to the total radiative forcing in 2100 for four RCPs (representative concentration pathways³) and at present day. The individual forcings comprise: the greenhouse gases carbon dioxide (CO2), methane (CH4), nitrous oxide (N₂O), ozone (O₃) and others; aerosols; and land use (LU). The right-hand panel shows the individual forcings relative to the total radiative forcing (i.e. RFx/RFtot, with RFx individual radiative forcings and RFtot total radiative forcing). Adapted from Collins et al. (2013: fig. 12.3).

Radiative forcing, W/m²

-1 0 1 2 3 4 5 6 7 8

Greenhouse gases Greenhouse gases

otheraerosolN2OCH4CO2O3LU

Percentage contribution of individual forcings to the total RCP8.5 (2100)

RCP6.0 (2100) RCP4.5 (2100) RCP2.6 (2100) Present day (2010)

-60 -40 -20 0 20 40 60 80 100

otheraerosolN2OCH4CO2O3LU

1.2

The Arctic climate context:

Past and future warming

To set the stage for the technical chapters that follow and to demonstrate the basis for concern about contemporary and future Arctic climate change, this section presents an updated estimate of recent Arctic warming and some illustrative maps of observed and potential future Arctic warming. It was not the intent to repeat the comprehensive analysis of Arctic climate undertaken as part of the AMAP SWIPA project (Overland et al. 2011b; Walsh et al. 2011). Previous analyses of long-term surface temperature data consistently show evidence of a strong amplification of warming in the Arctic region of about twice that of the rest of the world (Trenberth et al. 2007; Bekryaev et al. 2010; Overland et al. 2011a; Christensen et al. 2013; Jeffries and Richter-Menge 2013). Over the period 1950–2012, mean annual surface temperature (combined land and sea-surface temperatures) for the region north of 60°N has increased by about 1.6°C based on analysis of three data sets⁴ (see Fig. 1.2).

All three data sets indicate that warming has been strongest in spring (March–May), with mean increases in spring surface temperature of about 2°C (HadCRUT4: 1.95°C; GISS: 2.02°C;

MLOST: 1.83°C). Warming in autumn (Sept–Nov) and winter (Dec–Feb) has been only slightly less than that observed over the spring season (data not shown), while the weakest warming has been during summer (June–Aug). Maps of the seasonal warming trends for winter and summer (1950–2012) based on the NASA GISS data set (which has greater coverage over Arctic land areas than the other two data sets due to the method of interpolation used to fill in data between monitoring stations) are shown in Fig. 1.3. Winter mean temperature has risen 2.01°C (90% confidence interval: 1.26–2.76°C); while summer mean temperature has risen only 1.10°C (90% confidence interval:

0.68–1.45°C).

Looking forward, robust features of global model projections of climate change over the 21st century include a continuation of the observed large-scale trends for the Arctic, with strong regional warming about twice the global mean increase in annual surface temperature and with seasonally strongest warming in autumn and winter and weakest warming in summer (Collins et al. 2013).

For illustrative purposes, selected maps (Fig. 1.4) of average projected temperature changes, relative to the period 1986–

2005, have been drawn from the recently published Working Group I contribution to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC 2013c).

The selection illustrates projected changes over high latitude areas under two contrasting scenarios. With the RCP2.6 scenario, global temperature has stabilized or is declining from peak level (depending on the model) by the latter decades of the 21st century, with a mean rise in global average surface temperature, relative to the reference period, of 1.0°C over the period 2081–2100. With the RCP8.5 scenario, global

temperature is still rising by the end of the century, with a mean rise in global average surface temperature of 3.7°C over the period 2081–2100. While average future Arctic warming under the RCP2.6 scenario appears to be roughly comparable in magnitude to that observed since 1950, the high greenhouse gas emission assumptions of scenario RCP8.5 are projected to lead to dramatic changes in regional climate.

To further illustrate the potential for changes across the Arctic landscape under future warming, Fig. 1.5 shows how the locations around the circumpolar North, where the average annual air temperature is zero degrees centigrade, shifts northward under future climate change scenarios.

The diminishing size of the zero isotherm area over the progressively warmer climate scenarios, suggests regions where there is potential for permafrost degradation and release of carbon (as carbon dioxide or methane) from permafrost thaw, contributing to positive climate feedbacks. While permafrost distribution and thermal state are affected by other factors besides air temperature, and a wide variety of future permafrost states can be more rigorously diagnosed from climate models (e.g. Slater and Lawrence 2013), the figure is illustrative of the implications for natural ecosystem methane emissions (discussed in Ch. 3 and 4).

1.3

Report structure

There are significant challenges in acquiring and presenting an integrated understanding of the impact of changing methane emissions on Arctic climate. The approach taken in this report is to first present the group of chapters that deal with natural

-1.0 -0.5 0 0.5 1.0 1.5 2.0

Temperature anomaly, °C

2010 2000

1950 1960 1970 1980 1990

Mean 1961-1990

GISS MLOST HadCRUT4

Fig. 1.2 Annual average combined land and sea-surface temperature anomalies, 1950–2012 (relative to the mean over 1961–1990) for the area north of 60°N for three data sets: the Hadley Centre and Climate Research Unit dataset (HadCRUT4), the NASA Goddard Institute for Space Studies dataset (GISS), and the National Atmospheric and Oceanic Administration merged land-ocean surface temperature dataset (MLOST).

Trends (mean and 90% confidence intervals) over the 63-year period are 1.56°C (1.06–2.10°C) HadCRUT4; 1.73°C (1.15–2.32°C) GISS; and 1.46°C (1.06–1.83) MLOST.

4 This temperature analysis is based upon the Hadley Centre and Climate Research Unit dataset (HadCRUT4), the NASA Goddard Institute for Space Studies dataset (GISS), and the National Atmospheric and Oceanic Administration merged land-ocean surface temperature dataset (MLOST). Temperature anomalies in the three datasets are relative to different base periods. They are adjusted to a common base period 1961–1990 for time series plotting. The linear trend was computed using Mann-Kendall in combination with the Theil-Sen approach following Wang and Swail (2001) to account for auto-correlation in the time series.

For robust trend analysis, selection criteria were applied such that only sites with at least 50 years of data over the 63-year period, and at least six months of each year or two months of each season are used to compute annual or seasonal averages. Environment Canada, Climate Research Division, August, 2014.

0 0.5 1 1.25 1.5 1.75 2 2.5 3 4

HadCRUT4 annual GISS annual

GISS summer GISS winter

Observed temperature change, °C NOAA MLOST annual

Fig. 1.3 Observed Arctic warming, 1950–

2012, over the regions north of 60°N.

For the Hadley Centre and Climate Research Unit dataset (HadCRUT4) and the National Atmospheric and Oceanic Administration merged land-ocean surface temperature dataset (MLOST), temperatures are averaged over 5°×5°

grid boxes. The NASA Goddard Institute for Space Studies (GISS) data are averaged over 2°×2° grid boxes.

Projected temperature change, °C RCP2.6 Summer (JJA)

RCP8.5 Winter (DJF) RCP2.6 Winter (DJF)

RCP8.5 Summer (JJA)

Fig. 1.4 Projected change in surface air temperature over high-latitude areas for the period 2081–2100 relative to the reference period 1986–2005. Upper panels are median responses from 32 global climate models based on the RCP2.6 scenario, over the winter (DJF) and summer (JJA) seasons. Lower panels are based on 39 models using the RCP8.5 scenario. Adapted from IPCC Working Group I Fifth Assessment Report Annex I Supplemental Information (A1.SM2.6.21 and.23, and A1.SM8.5.21 and .23) (IPCC 2013c).

and anthropogenic methane sources, then present chapters that address atmospheric concentrations (as the atmosphere integrates emissions from all sources) and finally to present the results of modeling work that explicitly evaluates how changing sources will influence atmospheric concentrations and climate. As per the mandate to the Methane Expert Group, the scope of analysis is primarily Arctic-focused. While some AMAP assessments use a delineation of the Arctic region as defined by AMAP (land and marine areas north of the Arctic Circle, and north of 62°N in Asia and 60°N in North America, modified to include the marine areas north of the Aleutian chain, Hudson Bay, and parts of the North Atlantic Ocean including the Labrador Sea; AMAP 2011a, Fig. 1.1), for this assessment, no specific definition of an Arctic boundary was assumed and each chapter articulates boundaries suitable to the analysis within that chapter. Readers should note that while many of the chapters focus on the Arctic as a northern latitude region, where there is discussion of anthropogenic emissions of methane, the perspective is of Arctic nations as political entities, including all areas within their national borders. Given that methane is a global greenhouse gas, the technical chapters begin with an overview of the global methane budget (sources and sinks), and the role of methane as a greenhouse gas and climate forcer (Ch. 2). This provides essential background scientific information as well as the global context for understanding the subsequent chapters of the report, which are more Arctic focused. As Ch. 2 is provided for context, no key findings or conclusions are provided for that chapter.

Chapters 3 and 4 summarize current understanding of the natural processes that produce methane in Arctic environments and that may lead to enhanced emissions of methane from major terrestrial (Ch. 3) and marine (Ch. 4) sources. This work assesses the available published literature on these topics, drawing on both observational studies using flux measurements of methane to the atmosphere, and modeling studies. Emissions of methane from human activity are also changing and may also be contributing to recent changes in atmospheric methane levels. An assessment of available global methane emissions inventories is provided in Ch. 5 along with information specific to Arctic nations. In addition, Ch. 5 presents two scenarios of potential future anthropogenic methane emissions. One assumes no additional methane mitigation beyond existing legislation; the other is based on maximum technically feasible emission reductions with current abatement technologies. This scenario is global, but information specific to Arctic nations is extracted from the scenario.

Chapters 6 and 7 address the issue of how atmospheric concentrations of methane respond to changing emissions.

Chapter 6 presents an overview of the current atmospheric methane monitoring network over the Arctic region. Data from these sites are then analyzed and combined with previously published information to characterize trends and changes in atmospheric methane levels over time, on seasonal and longer time scales. Isotopic and trajectory analyses are explored as potential tools for detecting changes in methane emissions Fig. 1.5 The location of the zero degree near-surface air temperature isotherm for the historical 1996–

2005 period and the future 2081–

2100 period for the RCP2.6 and RCP8.5 scenarios. The zero degree isotherm is based on ensemble- mean annual air temperature simulated by 29 models (see below) that participated in the CMIP5 and averaged over the 10-year period centered on 2000 and 2090. Map created by Environment Canada’s Climate Research Division, December, 2014.

The 29 models from which results are used are BNU-ESM, CCSM4, CESM1-CAM5, CESM1- WACCM, CNRM-CM5, CSIRO- Mk3-6-0, CanESM2, EC-EARTH, FGOALS-g2, FIO-ESM, GFDL- CM3, GFDL-ESM2G, GFDL- ESM2M, GISS-E2-H, GISS-E2-R, HadGEM2-AO, HadGEM2-ES, IPSL-CM5A-LR, IPSL-CM5A- MR, MIROC-ESM, MIROC-ESM- CHEM, MIROC5, MPI-ESM-LR, MPI-ESM-MR, MRI-CGCM3, NorESM1-M, NorESM1-ME, bcc- csm1-1, and bcc-csm1-1-m.

0

0 0

Position of 0°C near-surface air temperature isotherm Historical (1996-2005)

RCP2.6 (2081-2100) RCP8.5 (2081-2100)

What is the potential benefit, in terms of reduced Arctic warming, of methane emissions mitigation by Arctic nations?

How does the magnitude of potential emission reductions from anthropogenic sources compare to potential changes in methane emissions from natural sources in the Arctic?

Chapters 3 & 4 What are the current and potential future natural emissions from the Arctic region?

Chapter 5

What are the current and potential future anthropogenic emissions of Arctic and non-Arctic nations?

Chapters 6 & 7

Are the current monitoring activities (of atmospheric concentrations and fluxes) sufficient to capture anticipated source changes?

Chapter 8

What is the historical and future Arctic climate response to changes in methane emissions, from Arctic and from global sources?

Various chapters What are the uncertainties in understanding the Arctic climate response to methane?

What are the current methane emissions from Arctic terrestrial and marine sources?

What are the controlling processes and factors that strongly influence natural emissions?

How may these emissions from natural sources in the Arctic change in the future?

What are the uncertainties or limitations in these estimates?

What are current global anthropogenic methane emissions, and those of Arctic nations?

How will the magnitude of emissions change in the future under different policy assumptions?

What percentage of global methane mitigation potential is controlled by Arctic Council nations?

What are the principal sources of uncertainty in these estimates of current and future anthropogenic emissions?

What are the trends and variability in Arctic methane concentrations and what are the primary drivers of this variability?

How much of a trend in atmospheric methane abundance can be detected with the current monitoring network?

Are emission estimates consistent with atmospheric concentrations?

Is there evidence of increasing Arctic methane emissions in the atmospheric observations?

What is the contribution of historical changes in global atmospheric methane to Arctic climate warming?

What impact will increasing atmospheric concentrations of methane have on climate and will Arctic nations have the ability to influence that impact through mitigation of anthropogenic methane emissions?

How will atmospheric methane concentrations change in response to potential changes in natural methane emissions and how do these changes compare to those that might result from mitigation of anthropogenic methane emissions?

Does the location of anthropogenic methane emissions matter?

Related to anthropogenic emissions characterization/

quantification/projection?

Related to natural emissions characterization/

quantification/projection from terrestrial and marine sources?

Related to climate response?

Related to measuring changes in atmospheric methane concentrations?

Table 1.1 Policy-relevant science questions guiding the work of the AMAP Methane Expert Group.

from different sources. Chapter 7 more explicitly integrates information on both atmospheric methane levels and emissions by taking a top-down inverse modeling approach to assess the extent to which changes in atmospheric methane in the Arctic can be explained and reconciled with estimates of natural and anthropogenic emissions in the Arctic. This information is also used to assess whether or not the atmospheric observations provide any indication of a trend in Arctic methane emissions.

In Ch. 8, the importance of past and potential future changes in methane emissions or concentrations on Arctic climate are discussed. In particular, the results of dedicated climate modeling experiments using the emission estimates from Ch. 3, 4, and 5 are presented, aimed at answering the overarching questions posed to the Methane Expert Group. Earth System Models are used to evaluate the benefit of anthropogenic methane emissions abatement in terms of reduced global and Arctic warming. Scenarios of natural emission change, founded on the analyses in Ch. 3 and 4, are used to calculate resulting changes in atmospheric methane concentration, allowing an estimate of warming resulting from such changes.

Each of the chapters in this report address a number of more specific policy-relevant science questions than the overarching questions presented at the end of Sect. 1.1. As a guide to the scope of work undertaken as part of this assessment, and to where readers can find information of particular interest, these questions are presented in Table 1.1. Key findings that respond to the questions in Table 1.1 are presented at the end of each chapter, along with recommendations for ongoing scientific work needed to address gaps in understanding. Chapter 9 presents a synthesis of these key findings and science recommendations.

The report concludes with a detailed summary of the strategies used for modelling the climate response. This annex is a common

contribution to the AMAP assessments on methane (the present report) and black carbon and ozone (AMAP 2015) and has been produced to facilitate an integrated understanding of the separate climate modelling exercises undertaken by the two AMAP expert groups on short-lived climate forcers (SLCFs).

Acknowledgments

The authors of this chapter are grateful for valuable comments on a draft of this chapter from John Walsh, University of Alaska Fairbanks, and from anonymous external reviewers. We are also grateful to Guilong Li, of Environment Canada’s Climate Research Division for his contributions to the development of Figs. 1.2 and 1.3.

2. The global methane budget and the role of methane in climate forcing

A: D P, A K

2.1

Background

Methane is emitted into the atmosphere from a large variety of sources and removed from the atmosphere predominately by chemical reactions. Since the lifetime of methane is approximately nine years (Prather et al. 2012), it is relatively well mixed throughout the troposphere and a simple global budget can be constructed as:

∆B₄ = Em – L Eq. 2.1 where ΔB_CH4 is the change in the global amount of methane in the atmosphere, E_m is the global total of emissions and L is the global total of methane losses. As discussed in more detail in Sect. 2.3, observations of the atmospheric concentration of methane and estimates of the rate of loss allow for tightly constrained estimates of ΔB_CH4 and L, respectively. Given these two terms and the associated uncertainties, the global total methane emission source can be constrained to approximately 10% (Prather et al. 2012). While the total emissions of methane are fairly well constrained, the division of the total among the individual sources and, in particular, the variability of individual sources and the contribution of changes in sources to the observed record of methane concentration is the subject of considerable research (e.g. Kai et al. 2011; Bergamaschi et al. 2013).

This chapter focuses primarily on the role of atmospheric chemistry in removing methane from the atmosphere, including current understanding of the magnitude and stability of this sink and the factors and mechanisms that influence it.

A discussion of the radiative forcing of climate by methane is also presented, including the influence of methane on tropospheric ozone, through atmospheric chemical processes, that accounts for a significant component of the radiative forcing due to methane. Only a brief overview of the current understanding of the magnitude of different methane sources is presented here. The reader is referred to subsequent chapters for a more in-depth discussion on this topic.

2.2

Overview of natural and

anthropogenic methane sources

The dominant sources of methane can be assigned to one of three categories – biogenic, thermogenic or pyrogenic.

Biogenic methane is produced by micro-organisms during the decomposition of organic carbon in anaerobic (low oxygen) environments (e.g. natural wetlands, flooded rice fields, landfills, termites, guts of ruminant animals) as well as in some natural- gas formations. Thermogenic methane, produced on geological timescales when deposits of organic material are exposed to high heat and pressure to form fossil fuels, is released (vented or leaked) when natural gas, oil and coal are extracted, processed and transported. Thermogenic methane may also enter the

atmosphere through naturally occurring pathways such as seeps and mud volcanoes. For an overview of biogenic and thermogenic sources see Cicerone and Oremland (1988). Pyrogenic methane is produced by the incomplete combustion of organic matter and includes sources such as biofuel burning, agricultural fires and wildfires (Andreae and Merlet 2001). A fourth category, abiogenic methane, results from chemical reactions involving inorganic carbon in the Earth’s crust. While the magnitude of emissions from abiogenic sources is very poorly known, it is not believed to be significant and is discussed further in Ch. 4.

For the purposes of this assessment, the three process-based categories of methane emission are further classified as either natural or anthropogenic sources. This is to delineate clearly both the ways in which anthropogenic activities have perturbed the methane cycle and the possible scope of mitigation measures to reduce anthropogenic methane emissions. Table 2.1 provides a recent synthesis of global methane source estimates. The estimates presented in Table 2.1 show anthropogenic activities to account for approximately 50% of the global total methane emissions, while other estimates suggest anthropogenic emissions may account for over 60% of the global total (e.g. Prather et al. 2012).

Global sources No.

studies Annual average emission, Tg CH4

Natural sources

Natural wetlands 3 217 (177–284)

Freshwater (lakes and rivers) 3 40 (8–73)

Wild animals (ruminants) 1 15 (15–15)

Wildfires 5 3 (1–5)

Termites 4 11 (2–22)

Geological 3 54 (33–75)

Marine 3 6 (2–9)^a

Permafrost (excl. lakes and wetlands) 1 1 (0–1)

Total 349 (238–492)

Anthropogenic sources

Rice cultivation 4 36 (33–40)

Domesticated animals (ruminants) 3 89 (87–94)

Landfills and waste 3 75 (67–90)

Biomass burning (incl. biofuels) 6 35 (32–39)

Fossil fuels 3 96 (85–105)

Total 331 (304–368)

a More recent estimates are provided in Ch. 4.

Table 2.1 Estimated annual average emissions for the major methane sources over the period 2000–2009 from Kirschke et al. (2013). The estimate for each source is calculated as the mean of a variable number of individual studies for each source. The range (in brackets) is defined as the maximum and minimum values from the studies reviewed.

The estimates presented in Table 2.1 are derived from

‘bottom-up’ studies. These are studies where the emissions from a sample of a particular source type are estimated, often based on measurements taken in the real world. The measured emissions from the sample are then extrapolated to derive the global total using estimates of the global extent of the source.

As such, the global total of the independently estimated sources is not constrained by estimates of the global total of emissions.

Indeed, the sum of the best-guess estimates for each source from the bottom-up studies yields a global total emission of 680 Tg CH₄/y (Table 2.1), which falls outside the uncertainty range of the ‘top-down’ estimate of global total emissions of 554 ± 56 Tg CH₄/y derived from the methane abundance and estimates of atmospheric lifetime (Prather et al. 2012).

Comparison with the top-down estimate suggests global total methane emissions towards the lower end of the range given by the bottom-up studies, where the two estimates overlap, however, the comparison does not provide information on where the net overestimate in the bottom-up studies may originate.

In addition to the bottom-up estimates and the constraints from the atmospheric lifetime and observed concentration, some additional methods to estimate emissions have been used, including atmospheric inversion (discussed further in Ch. 7) and the analysis of methane isotopes (discussed further in Ch. 6). These methods provide additional information on the broad regional (continental-scale) distribution of sources or source categories (biogenic versus thermogenic).

The significant contribution of anthropogenic sources to the global total of methane emissions provides an indication of the degree to which anthropogenic activities have perturbed the methane budget. Ice cores provide robust evidence that atmospheric concentrations⁵ were around 720 ppb in 1750 (Ciais et al. 2013), while atmospheric measurements show a global average methane concentration of 1819 ppb in 2012 (WMO 2014). Most of the increase in methane is believed to be due to increased emissions resulting from anthropogenic activities, notably rice cultivation, ruminant livestock, landfills and fossil fuel extraction and use.

In terms of the more recent past, the rate of increase in the atmospheric concentration of methane has decreased since the mid-1980s, approaching a near-zero growth rate over 1999–2006 (see Fig. 2.1), before resuming a slower increase from 2007 onwards. The considerable year-to-year variability in the rate of increase in the atmospheric concentration over the 1990s has been attributed to the eruption of Mt. Pinatubo and the collapse of the Soviet Union in 1991/92 (Dlugokencky et al. 1996; Bousquet et al. 2006) and a strong El Niño in 1997/98 (Bousquet et al. 2006). The longer-term stability of the atmospheric methane concentration over 1999–2006 indicates a rough balance between sources and sinks during this period.

Current understanding of methane sinks suggests these are fairly stable with time (discussed further in Sect. 2.3) and argues for changes in methane sources as the main reason behind recent changes in methane concentration growth rates, although attributing the changes in emissions to particular sources has proved challenging. Some studies have suggested that methane

emissions from fossil fuel extraction and distribution decreased by 10 to 30 Tg CH₄/y between the 1980s and 2000s, with much of this decrease occurring before 2000 (Aydin et al. 2011;

Simpson et al. 2012). It has also been suggested that decreases in microbial emissions, particularly due to changes in the practice of rice cultivation, could be responsible for a ~15 Tg CH₄/y decrease over roughly the same time period (Kai et al. 2011).

Uncertainty in the observations used to derive the decrease in microbial emissions (Levin et al. 2012) and the possibility of an offsetting increase in microbial emissions from sources such as natural wetlands or ruminants (Kirschke et al. 2013) further complicate an understanding of the causes of the stabilization of methane concentrations in the early 2000s.

A rise in atmospheric methane concentration resumed in 2007, albeit at a slower rate than in the 1980s. Figure 2.2 presents the latitudinally-resolved growth rate in the near-surface concentration of methane over 2000 to 2014. Clearly evident are strong increases in methane in 2007 at high latitudes in the northern hemisphere and in tropical latitudes in both 2007/08 and 2010/11. The increases are believed to be driven by increased emissions from wetlands due to year-to-year variability in meteorological conditions (Bousquet et al. 2011).

The contribution of these anomalous years to the resumption of growth in the global-average methane concentration, including the possibility of additional contributions from changes in anthropogenic emissions, are discussed in Ch. 5 and 6.

5 Throughout this report the atmospheric concentration of methane will be given as the volume mixing ratio, also referred to as the molar mixing ratio, defined as the number density of methane relative to the number density of dry air. For simplicity, the term ‘concentration’ is used throughout.

Fig. 2.1 The global-average methane concentration derived from surface observations by the US National Oceanic and Atmospheric Administration (NOAA) Cooperative Air Sampling Network (upper panel; the dashed red line is the trend line fitted to the deseasonalized data) and (lower panel) the annual rate of change in methane concentration (full red line) calculated from the dashed line in the upper panel, along with the 1-sigma uncertainty (hatched red lines). Updated from Dlugokencky et al. (2011).

1775

1725

1675 CH4 in air, ppb

Global average

Methane growth 15

10 5 0 -5

1985 1990

Rate of change in CH4, ppb/y

1995 2000 2005 2010